Combined use of waist and thigh circumference to identify high‐risk, abdominally obese HIV+ patients

Abstract

Background: We examined whether the combination of waist (WC) and thigh (ThC) circumference improves the prediction of visceral adipose tissue (VAT) over WC and ThC independently in HIV-infected men and women after correction for age. We also examined the independent associations between VAT, and the combination of WC and ThC with metabolic risk factors, metabolic syndrome, type 2 diabetes mellitus (T2DM) and prior cardiovascular events in HIVinfected individuals. Methods: Consecutive patients attending the metabolic clinic of the University of Modena in Italy between 2005 and 2009 were recruited in this cross-sectional study. Total and regional fat mass and lean mass were quantified using DEXA. A single CT image was taken for quantification of VAT and CAC. Prior cardiovascular events which occurred within a 5-year period of the clinical evaluation were analysed. A cross-fold test was used to explore different models in the ability to predict VAT in order to build an algorithm for VAT estimation (e-VAT). Regression analysis were performed to determine the univariate and multivariate relations between WC, ThC, and age with VAT. A comparison of beta coefficients for VAT and e-VAT to predict cardio-metabolic risk and events were performed using multivariable regression models after correction for BMI and age. Results: 2322 HIV-infected patients were recruited: median duration of HIV infection was 182 months (IQR 126-236); median nadir and current CD4 were 172 (IQR 68-262) and 515.5 (IQR 369-700) and 75% of them had undetectable HIV1-VL. In this abstract only the results of men will be presented. Men (n=1481) had a mean age of 45.997.3 years, a BMI of 24.1 ± 3.8 kg/m2, a WC of 88.0910.1 cm and a ThC of 47.894.3 cm. e-VAT algorithm for men was: (5.44∗WC) - (1.35∗ThC) - (1.70∗age) -348.1 In men, at multivariable regression models after correction for BMI and age, e-VAT was concordant to VAT in predicting HOMA, MetS Risk, prior cardiovascular events (OR=1.01), was better than VAT in predicting T2DM (OR=1.00) and CAC-10 (OR=1.01) but was worse than VAT in predicting TC/HDL and TG. Discussion: We confirm that ThC is inversely associated to VAT after correction for WC. e-VAT is a sensitive tool to predict VAT more accurately than WC and ThC independently. e-VAT proved to predict cardio-metabolic risks and events in men and women, qualifying this variable for a potential clinical use.